Bug 457957

Summary: FEAT: add oprofile testing to hts
Product: [Retired] Red Hat Hardware Certification Program Reporter: Rob Landry <rlandry>
Component: Test Suite (tests)Assignee: Greg Nichols <gnichols>
Status: CLOSED ERRATA QA Contact: Lawrence Lim <llim>
Severity: medium Docs Contact:
Priority: medium    
Version: 5.2CC: benl, gnichols, jturner, rlandry, rousseau, tburke, tools-bugs, tyan, wcohen, ykun
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-01-27 22:57:47 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
profile.py - a simple oprofile test
none
command.py patch for simpler "grep"-style searches
none
profile.py - a simple oprofile test
none
profiler.py patch for ia64 hang - also adds calls to opreport none

Description Rob Landry 2008-08-05 18:58:19 UTC
Description of problem:

Tim inquired about adding oprofile testing to the hwcert test suite (this was the begining of May), Will provided some insight into the location of existing testings which could be leveraged to avoiding needing to write everything from scratch (pasted below).

This should be considered for the hts 5.3 release; additionally we'll need to consider the policy impacts, do all systems require it, if so do all systems require the specialized processor support or is the time mech. enough, if just the timer mech. works is a kbase required, etc.


Additional info:

> > >> The tests available for oprofile are rather simple "smoke" tests to
> > >> see that oprofile can be started and stopped. and whether it can
> > >> collect data. You can see the tests on the single workflow for the
> > >> oprofile package, /tools/oprofile/testsuite.
> > >>
> > >> Note that oprofile has a fall back timer interrupt mechanism. This is
> > >> used if oprofile cannot identify the performance monitoring hardware
> > >> or cannot use the processor's performance monitoring hardware. Some
> > >> users of oprofile might be disappointed that oprofile uses the
> > >> fallback timer interrupt mechanism and push for oprofile to support
> > >> the processors performance monitoring hardware (e.g. IBM power
> > >> processors). However, the tests pass if oprofile is able to use the
> > >> either the performance monitoring hardware OR the timer interrupt
> > >> mechanism. This would catch the Goldman Sach's issue, but not IBM's
> > >> comment that particular processor's performance monitoring hardware
> > >> is supported.

Comment 1 Greg Nichols 2008-11-26 16:52:29 UTC
Created attachment 324760 [details]
profile.py - a simple oprofile test


This test runs oprofile and its daemon, and verifies the presence of the oprofile module, filesystem, and daemon process.

Comment 2 Greg Nichols 2008-11-26 16:55:11 UTC
Created attachment 324761 [details]
command.py patch for simpler "grep"-style searches



This patch improves the Command class for support of simple search commands.

Comment 3 Greg Nichols 2008-11-26 16:57:06 UTC
Created attachment 324762 [details]
profile.py - a simple oprofile test

Comment 6 Greg Nichols 2008-12-08 18:20:16 UTC
Created attachment 326156 [details]
profiler.py patch for ia64 hang - also adds calls to opreport


This patch is against R4, and contains fixes for two different hanging problems on ia64, and also adds calls to opreport to check reporting.

Comment 10 Yan Tian 2008-12-30 11:31:38 UTC
Verified `hts print plan` listed profiler test and `hts run --test profiler` worked well with hts-5.3-12.

Comment 13 errata-xmlrpc 2009-01-27 22:57:47 UTC
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHBA-2009-0047.html